Deepfake Fraud Tools Are Lagging Behind Expectations

Deepfake-generating software has been evolving from a toy into a legitimate fraud threat, but evidence suggests that progression is happening more slowly than predicted, according to World Economic Forum (WEF) researchers.

They reviewed 17 open source and commercially available deepfake programs that were available online between July 2024 and April 2025 and found the majority to be relatively cheap and superficial, marketed for social media or entertainment, with a minority showing more advanced features that could enable identity fraud.

Tom Cross, head of threat research at GetReal Security, warned that “they’re available on the black market, one hundred and fifty to two-hundred bucks per account,” and that such accounts are being bought by people engaged in money laundering, with threat actors able to obtain KYC-validated bank accounts using deepfakes.

Even among tools that pass the eye test, the study noted many can struggle with live KYC checks, though some can inject fake imagery into video feeds and use motion-capture to mimic subtle movements. Yet defenders remain ahead of attackers, with experts arguing that detection software and multi-layered verification approaches undermine the practical use of many deepfakes in real-world fraud.